arXiv subjects

Can Cui

Publications and source records attributed to Can Cui.

At least 55 records · Page 3Linked to original sources

arXiv2025

Speaker Embeddings to Improve Tracking of Intermittent and Moving Speakers

Speaker tracking methods often rely on spatial observations to assign coherent track identities over time. This raises limits in scenarios with intermittent and moving speakers, i.e., speakers that may change position when they are inactive, thus leading to discontinuous spatial trajectories. This paper proposes to investigate the use of speaker embeddings, in a simple solution to this issue. We propose to perform identity reassignment post-tracking, using speaker embeddings. We leverage trajectory-related information provided by an initial tracking step and multichannel audio signal. Beamforming is used to enhance the signal towards the speakers' positions in order to compute speaker embeddings. These are then used to assign new track identities based on an enrollment pool. We evaluate the performance of the proposed speaker embedding-based identity reassignment method on a dataset where speakers change position during inactivity periods. Results show that it consistently improves the identity assignment performance of neural and standard tracking systems. In particular, we study the impact of beamforming and input duration for embedding extraction.

eess.AS↗

arXiv2025

Joint UAV Trajectory Planning and LEO Satellite Selection for Data Offloading in Space-Air-Ground Integrated Networks

With the development of low earth orbit (LEO) satellites and unmanned aerial vehicles (UAVs), the space-air-ground integrated network (SAGIN) becomes a major trend in the next-generation networks. However, due to the instability of heterogeneous communication and time-varying characteristics of SAGIN, it is challenging to meet the remote Internet of Things (IoT) demands for data collection and offloading. In this paper, we investigate a two-phase hierarchical data uplink model in SAGIN. Specifically, UAVs optimize trajectories to enable efficient data collection from IoT devices, and then they transmit the data to LEO satellites with computing capabilities for further processing. The problem is formulated to minimize the total energy consumption for IoT devices, UAVs, and LEO satellites. Since the problem is in the form of mixed-integer nonlinear programming and intractable to solve directly, we decompose it into two phases. In the IoT-UAV phase, we design the algorithm to jointly optimize the IoT pairing, power allocation, and UAVs trajectories. Considering the high dynamic characteristics of LEO satellites, a real-time LEO satellite selection mechanism joint with the Satellite Tool Kit is proposed in the UAV-LEO phase. Finally, simulation results show the effectiveness of the proposed algorithms, with about 10% less energy consumption compared with the benchmark algorithm.

cs.SI↗

arXiv2025

A Hierarchical Test Platform for Vision Language Model (VLM)-Integrated Real-World Autonomous Driving

Vision-Language Models (VLMs) have demonstrated notable promise in autonomous driving by offering the potential for multimodal reasoning through pretraining on extensive image-text pairs. However, adapting these models from broad web-scale data to the safety-critical context of driving presents a significant challenge, commonly referred to as domain shift. Existing simulation-based and dataset-driven evaluation methods, although valuable, often fail to capture the full complexity of real-world scenarios and cannot easily accommodate repeatable closed-loop testing with flexible scenario manipulation. In this paper, we introduce a hierarchical real-world test platform specifically designed to evaluate VLM-integrated autonomous driving systems. Our approach includes a modular, low-latency on-vehicle middleware that allows seamless incorporation of various VLMs, a clearly separated perception-planning-control architecture that can accommodate both VLM-based and conventional modules, and a configurable suite of real-world testing scenarios on a closed track that facilitates controlled yet authentic evaluations. We demonstrate the effectiveness of the proposed platform`s testing and evaluation ability with a case study involving a VLM-enabled autonomous vehicle, highlighting how our test framework supports robust experimentation under diverse conditions.

cs.RO↗

arXiv2025

RationalVLA: A Rational Vision-Language-Action Model with Dual System

A fundamental requirement for real-world robotic deployment is the ability to understand and respond to natural language instructions. Existing language-conditioned manipulation tasks typically assume that instructions are perfectly aligned with the environment. This assumption limits robustness and generalization in realistic scenarios where instructions may be ambiguous, irrelevant, or infeasible. To address this problem, we introduce RAtional MAnipulation (RAMA), a new benchmark that challenges models with both unseen executable instructions and defective ones that should be rejected. In RAMA, we construct a dataset with over 14,000 samples, including diverse defective instructions spanning six dimensions: visual, physical, semantic, motion, safety, and out-of-context. We further propose the Rational Vision-Language-Action model (RationalVLA). It is a dual system for robotic arms that integrates the high-level vision-language model with the low-level manipulation policy by introducing learnable latent space embeddings. This design enables RationalVLA to reason over instructions, reject infeasible commands, and execute manipulation effectively. Experiments demonstrate that RationalVLA outperforms state-of-the-art baselines on RAMA by a 14.5% higher success rate and 0.94 average task length, while maintaining competitive performance on standard manipulation tasks. Real-world trials further validate its effectiveness and robustness in practical applications. Our project page is https://irpn-eai.github.io/RationalVLA.

cs.RO↗

arXiv2025

IRS: Incremental Relationship-guided Segmentation for Digital Pathology

Continual learning is rapidly emerging as a key focus in computer vision, aiming to develop AI systems capable of continuous improvement, thereby enhancing their value and practicality in diverse real-world applications. In healthcare, continual learning holds great promise for continuously acquired digital pathology data, which is collected in hospitals on a daily basis. However, panoramic segmentation on digital whole slide images (WSIs) presents significant challenges, as it is often infeasible to obtain comprehensive annotations for all potential objects, spanning from coarse structures (e.g., regions and unit objects) to fine structures (e.g., cells). This results in temporally and partially annotated data, posing a major challenge in developing a holistic segmentation framework. Moreover, an ideal segmentation model should incorporate new phenotypes, unseen diseases, and diverse populations, making this task even more complex. In this paper, we introduce a novel and unified Incremental Relationship-guided Segmentation (IRS) learning scheme to address temporally acquired, partially annotated data while maintaining out-of-distribution (OOD) continual learning capacity in digital pathology. The key innovation of IRS lies in its ability to realize a new spatial-temporal OOD continual learning paradigm by mathematically modeling anatomical relationships between existing and newly introduced classes through a simple incremental universal proposition matrix. Experimental results demonstrate that the IRS method effectively handles the multi-scale nature of pathological segmentation, enabling precise kidney segmentation across various structures (regions, units, and cells) as well as OOD disease lesions at multiple magnifications. This capability significantly enhances domain generalization, making IRS a robust approach for real-world digital pathology applications.

eess.IV↗

arXiv2025

QUART-Online: Latency-Free Large Multimodal Language Model for Quadruped Robot Learning

This paper addresses the inherent inference latency challenges associated with deploying multimodal large language models (MLLM) in quadruped vision-language-action (QUAR-VLA) tasks. Our investigation reveals that conventional parameter reduction techniques ultimately impair the performance of the language foundation model during the action instruction tuning phase, making them unsuitable for this purpose. We introduce a novel latency-free quadruped MLLM model, dubbed QUART-Online, designed to enhance inference efficiency without degrading the performance of the language foundation model. By incorporating Action Chunk Discretization (ACD), we compress the original action representation space, mapping continuous action values onto a smaller set of discrete representative vectors while preserving critical information. Subsequently, we fine-tune the MLLM to integrate vision, language, and compressed actions into a unified semantic space. Experimental results demonstrate that QUART-Online operates in tandem with the existing MLLM system, achieving real-time inference in sync with the underlying controller frequency, significantly boosting the success rate across various tasks by 65%. Our project page is https://quart-online.github.io.

cs.RO↗

arXiv2025

Distributionally Robust Optimization for Aerial Multi-access Edge Computing via Cooperation of UAVs and HAPs

With an extensive increment of computation demands, the aerial multi-access edge computing (MEC), mainly based on unmanned aerial vehicles (UAVs) and high altitude platforms (HAPs), plays significant roles in future network scenarios. In detail, UAVs can be flexibly deployed, while HAPs are characterized with large capacity and stability. Hence, in this paper, we provide a hierarchical model composed of an HAP and multi-UAVs, to provide aerial MEC services. Moreover, considering the errors of channel state information from unpredictable environmental conditions, we formulate the problem to minimize the total energy cost with the chance constraint, which is a mixed-integer nonlinear problem with uncertain parameters and intractable to solve. To tackle this issue, we optimize the UAV deployment via the weighted K-means algorithm. Then, the chance constraint is reformulated via the distributionally robust optimization (DRO). Furthermore, based on the conditional value-at-risk mechanism, we transform the DRO problem into a mixed-integer second order cone programming, which is further decomposed into two subproblems via the primal decomposition. Moreover, to alleviate the complexity of the binary subproblem, we design a binary whale optimization algorithm. Finally, we conduct extensive simulations to verify the effectiveness and robustness of the proposed schemes by comparing with baseline mechanisms.

cs.DC↗

arXiv2025

OpenHelix: A Short Survey, Empirical Analysis, and Open-Source Dual-System VLA Model for Robotic Manipulation

Dual-system VLA (Vision-Language-Action) architectures have become a hot topic in embodied intelligence research, but there is a lack of sufficient open-source work for further performance analysis and optimization. To address this problem, this paper will summarize and compare the structural designs of existing dual-system architectures, and conduct systematic empirical evaluations on the core design elements of existing dual-system architectures. Ultimately, it will provide a low-cost open-source model for further exploration. Of course, this project will continue to update with more experimental conclusions and open-source models with improved performance for everyone to choose from. Project page: https://openhelix-robot.github.io/.

cs.RO↗

arXiv2025

Rossby wave instability in weakly ionized protoplanetary disks. II. radial B-fields

Building on our first paper in this series, we investigate the impact of radial magnetic fields and non-ideal magnetohydrodynamic (MHD) effects - specifically, Ohmic resistivity, Hall drift, and ambipolar diffusion - on RWI unstable modes. The presence of a radial field is linked to the disk's vertical shear and vertical magnetic field. We perform radially global linear analyses and utilize the spectral code \textsc{Dedalus} to solve the matrix eigenvalue problems. Our findings reveal that radial fields exhibit behavior similar to vertical fields. In the ideal MHD limit, radial fields enhance the effect of vertical fields in reducing growth rates, with significant reductions starting at relatively weak field strengths, around $β\sim 10^3 - 10^4$, which are relevant to protoplanetary disks. In the non-ideal MHD limit, all three non-ideal effects, when sufficiently strong, cause the growth rates to closely resemble those observed in hydrodynamic models.

astro-ph.EP↗

arXiv2025

Dust Growth in ALMA Rings: II. Dusty Rossby Wave Instability

Annular substructures serve as ideal venues for planetesimal formation. In this series, we investigate the linear stage of dust growth within rings. The first paper examines the global streaming instability, while this study focuses on the dusty Rossby wave instability (DRWI). We perform a linear analysis of the two-fluid equations on a background pressure bump, representing annular substructures. The spectral code \textsc{Dedalus} is used to solve the linear eigenvalue problem. We identify two distinct DRWI modes: Type I, which originates from dust-modified gas RWI, and Type II, which results from dust-gas coupling. These modes never coexist for a given azimuthal wavenumber $\ky$, but transition between each other as $\ky$ varies. Type I modes are driven by the advection of background vorticity, whereas Type II modes involve two primary waves: Rossby waves, driven by advection, and thin waves, driven by dust-gas drag. Finally, we assess the relevance of DRWI in ALMA rings using DSHARP sources. Our findings suggest that Type I modes could explain the absence of azimuthal asymmetries in many ALMA disks, whereas Type II modes are entirely absent in all eight observed rings, implying that unresolved narrow rings or alternative mechanisms may play a role in dust growth within annular substructures.

astro-ph.EP↗

arXiv2025

Enhancing Adversarial Transferability via Component-Wise Transformation

Deep Neural Networks (DNNs) are highly vulnerable to adversarial examples, which pose significant challenges in security-sensitive applications. Among various adversarial attack strategies, input transformation-based attacks have demonstrated remarkable effectiveness in enhancing adversarial transferability. However, existing methods still perform poorly across different architectures, even though they have achieved promising results within the same architecture. This limitation arises because, while models of the same architecture may focus on different regions of the object, the variation is even more pronounced across different architectures. Unfortunately, current approaches fail to effectively guide models to attend to these diverse regions. To address this issue, this paper proposes a novel input transformation-based attack method, termed Component-Wise Transformation (CWT). CWT applies interpolation and selective rotation to individual image blocks, ensuring that each transformed image highlights different target regions, thereby improving the transferability of adversarial examples. Extensive experiments on the standard ImageNet dataset show that CWT consistently outperforms state-of-the-art methods in both attack success rates and stability across CNN- and Transformer-based models.

cs.CV↗

arXiv2025

Dataset Distillation in Medical Imaging: A Feasibility Study

Data sharing in the medical image analysis field has potential yet remains underappreciated. The aim is often to share datasets efficiently with other sites to train models effectively. One possible solution is to avoid transferring the entire dataset while still achieving similar model performance. Recent progress in data distillation within computer science offers promising prospects for sharing medical data efficiently without significantly compromising model effectiveness. However, it remains uncertain whether these methods would be applicable to medical imaging, since medical and natural images are distinct fields. Moreover, it is intriguing to consider what level of performance could be achieved with these methods. To answer these questions, we conduct investigations on a variety of leading data distillation methods, in different contexts of medical imaging. We evaluate the feasibility of these methods with extensive experiments in two aspects: 1) Assess the impact of data distillation across multiple datasets characterized by minor or great variations. 2) Explore the indicator to predict the distillation performance. Our extensive experiments across multiple medical datasets reveal that data distillation can significantly reduce dataset size while maintaining comparable model performance to that achieved with the full dataset, suggesting that a small, representative sample of images can serve as a reliable indicator of distillation success. This study demonstrates that data distillation is a viable method for efficient and secure medical data sharing, with the potential to facilitate enhanced collaborative research and clinical applications.

eess.IV↗

arXiv2025

GLAM: Glomeruli Segmentation for Human Pathological Lesions using Adapted Mouse Model

Moving from animal models to human applications in preclinical research encompasses a broad spectrum of disciplines in medical science. A fundamental element in the development of new drugs, treatments, diagnostic methods, and in deepening our understanding of disease processes is the accurate measurement of kidney tissues. Past studies have demonstrated the viability of translating glomeruli segmentation techniques from mouse models to human applications. Yet, these investigations tend to neglect the complexities involved in segmenting pathological glomeruli affected by different lesions. Such lesions present a wider range of morphological variations compared to healthy glomerular tissue, which are arguably more valuable than normal glomeruli in clinical practice. Furthermore, data on lesions from animal models can be more readily scaled up from disease models and whole kidney biopsies. This brings up a question: ``\textit{Can a pathological segmentation model trained on mouse models be effectively applied to human patients?}" To answer this question, we introduced GLAM, a deep learning study for fine-grained segmentation of human kidney lesions using a mouse model, addressing mouse-to-human transfer learning, by evaluating different learning strategies for segmenting human pathological lesions using zero-shot transfer learning and hybrid learning by leveraging mouse samples. From the results, the hybrid learning model achieved superior performance.

eess.IV↗

arXiv2025

Assessment of Cell Nuclei AI Foundation Models in Kidney Pathology

Cell nuclei instance segmentation is a crucial task in digital kidney pathology. Traditional automatic segmentation methods often lack generalizability when applied to unseen datasets. Recently, the success of foundation models (FMs) has provided a more generalizable solution, potentially enabling the segmentation of any cell type. In this study, we perform a large-scale evaluation of three widely used state-of-the-art (SOTA) cell nuclei foundation models (Cellpose, StarDist, and CellViT). Specifically, we created a highly diverse evaluation dataset consisting of 2,542 kidney whole slide images (WSIs) collected from both human and rodent sources, encompassing various tissue types, sizes, and staining methods. To our knowledge, this is the largest-scale evaluation of its kind to date. Our quantitative analysis of the prediction distribution reveals a persistent performance gap in kidney pathology. Among the evaluated models, CellViT demonstrated superior performance in segmenting nuclei in kidney pathology. However, none of the foundation models are perfect; a performance gap remains in general nuclei segmentation for kidney pathology.

eess.IV↗

arXiv2025

Rossby wave instability in weakly ionized protoplanetary disks. I. azimuthal or vertical B-fields

Rossby wave instability (RWI) is considered the underlying mechanism to crescent-shaped azimuthal asymmetries, discovered in (sub-)millimeter dust continuum of many protoplanetary disks. Previous works on linear theory were conducted in the hydrodynamic limit. Nevertheless, protoplanetary disks are likely magnetized and weakly ionized. We examine the influence of magnetic fields and non-ideal magnetohydrodynamic (MHD) effects - namely, Ohmic resistivity, Hall drift, and ambipolar diffusion - on the RWI unstable modes. We perform radially global linear analyses, employing constant azimuthal ($B_ϕ$) or vertical ($B_z$) background magnetic fields. It is found that, in the ideal MHD regime, magnetism can either enhance or diminish RWI growth. Strong non-ideal MHD effects cause RWI growth rates to recover hydrodynamic results. The sign of Hall Elsässer number subtly complicates the results, and vertical wavenumbers generically diminish growth rates.

astro-ph.EP↗

arXiv2024

Score and Distribution Matching Policy: Advanced Accelerated Visuomotor Policies via Matched Distillation

Visual-motor policy learning has advanced with architectures like diffusion-based policies, known for modeling complex robotic trajectories. However, their prolonged inference times hinder high-frequency control tasks requiring real-time feedback. While consistency distillation (CD) accelerates inference, it introduces errors that compromise action quality. To address these limitations, we propose the Score and Distribution Matching Policy (SDM Policy), which transforms diffusion-based policies into single-step generators through a two-stage optimization process: score matching ensures alignment with true action distributions, and distribution matching minimizes KL divergence for consistency. A dual-teacher mechanism integrates a frozen teacher for stability and an unfrozen teacher for adversarial training, enhancing robustness and alignment with target distributions. Evaluated on a 57-task simulation benchmark, SDM Policy achieves a 6x inference speedup while having state-of-the-art action quality, providing an efficient and reliable framework for high-frequency robotic tasks.

cs.RO↗

arXiv2024

Robust Fiber Orientation Distribution Function Estimation Using Deep Constrained Spherical Deconvolution for Diffusion MRI

Diffusion-weighted magnetic resonance imaging (DW-MRI) is a critical imaging method for capturing and modeling tissue microarchitecture at a millimeter scale. A common practice to model the measured DW-MRI signal is via fiber orientation distribution function (fODF). This function is the essential first step for the downstream tractography and connectivity analyses. With recent advantages in data sharing, large-scale multi-site DW-MRI datasets are being made available for multi-site studies. However, measurement variabilities (e.g., inter- and intra-site variability, hardware performance, and sequence design) are inevitable during the acquisition of DW-MRI. Most existing model-based methods (e.g., constrained spherical deconvolution (CSD)) and learning based methods (e.g., deep learning (DL)) do not explicitly consider such variabilities in fODF modeling, which consequently leads to inferior performance on multi-site and/or longitudinal diffusion studies. In this paper, we propose a novel data-driven deep constrained spherical deconvolution method to explicitly constrain the scan-rescan variabilities for a more reproducible and robust estimation of brain microstructure from repeated DW-MRI scans. Specifically, the proposed method introduces a new 3D volumetric scanner-invariant regularization scheme during the fODF estimation. We study the Human Connectome Project (HCP) young adults test-retest group as well as the MASiVar dataset (with inter- and intra-site scan/rescan data). The Baltimore Longitudinal Study of Aging (BLSA) dataset is employed for external validation. From the experimental results, the proposed data-driven framework outperforms the existing benchmarks in repeated fODF estimation. The proposed method is assessing the downstream connectivity analysis and shows increased performance in distinguishing subjects with different biomarkers.

cs.CV↗

arXiv2024

Glo-In-One-v2: Holistic Identification of Glomerular Cells, Tissues, and Lesions in Human and Mouse Histopathology

Segmenting glomerular intraglomerular tissue and lesions traditionally depends on detailed morphological evaluations by expert nephropathologists, a labor-intensive process susceptible to interobserver variability. Our group previously developed the Glo-In-One toolkit for integrated detection and segmentation of glomeruli. In this study, we leverage the Glo-In-One toolkit to version 2 with fine-grained segmentation capabilities, curating 14 distinct labels for tissue regions, cells, and lesions across a dataset of 23,529 annotated glomeruli across human and mouse histopathology data. To our knowledge, this dataset is among the largest of its kind to date.In this study, we present a single dynamic head deep learning architecture designed to segment 14 classes within partially labeled images of human and mouse pathology data. Our model was trained using a training set derived from 368 annotated kidney whole-slide images (WSIs) to identify 5 key intraglomerular tissues covering Bowman's capsule, glomerular tuft, mesangium, mesangial cells, and podocytes. Additionally, the network segments 9 glomerular lesion classes including adhesion, capsular drop, global sclerosis, hyalinosis, mesangial lysis, microaneurysm, nodular sclerosis, mesangial expansion, and segmental sclerosis. The glomerulus segmentation model achieved a decent performance compared with baselines, and achieved a 76.5 % average Dice Similarity Coefficient (DSC). Additional, transfer learning from rodent to human for glomerular lesion segmentation model has enhanced the average segmentation accuracy across different types of lesions by more than 3 %, as measured by Dice scores. The Glo-In-One-v2 model and trained weight have been made publicly available at https: //github.com/hrlblab/Glo-In-One_v2.

eess.IV↗